The Determinants of the Implementation Speed of Cohesion Projects in Europe: A Regional and Project-level Analysis
- Autori: Gambina, D.; Mazzola, F.
- Anno di pubblicazione: 2025
- Tipologia: Articolo in rivista
- OA Link: http://hdl.handle.net/10447/687723
Abstract
Time overruns of public works hinder the impact of investment policies and spill over negatively on aggregate prosperity. Particularly for sustainability-related projects, postponing operations may hamper the environmental objectives. This paper highlights the regional dimension of the problem and evaluates the hypothesis that the regional conditions in which the projects are implemented affect time efficiency of EU cohesion policy. We deal with interventions of the 2014–2020 programming period targeted to sustainable infrastructures and energy saving of SMEs and big companies by extracting the relevant information from Kohesio database. We apply linear regression models and logistic regressions as well as survival techniques to support our results. The evaluation is carried out at two scales. In the NUTS-2 framework, the duration is explained by proxies of territorial context such as development level, urbanization and institutional quality calculated at the regional level. At a micro-level, we scrutinize the effects of regional conditions, in comparison with those of project-specific features, also across categories of interventions. According to our findings, urbanization and institutional capacity turn out to drive timely completion while per capita GDP, capturing project dimensionality and complexity, does not. Moreover, we find that time efficiency increases monotonically as we approach the end of the programming period and that national cofinancing lengthens the duration, especially of the incentives to large firms. Our assessment enriches the debate on regional policy effectiveness by addressing a topic that has not been thoroughly empirically tested at the supranational level and offering policy suggestions relevant to the 2021–2027 programming.